Pediatr Radiol DOI 10.1007/s00247-017-3939-1
ORIGINAL ARTICLE
Non-central peripherally inserted central catheters in neonatal intensive care: complication rates and longevity of catheters relative to tip position Bernard Goldwasser 1 Robert M. Angert 6
&
Catalina Baia 2 & Mimi Kim 3 & Benjamin H. Taragin 4,5 &
Received: 28 February 2017 / Revised: 8 June 2017 / Accepted: 30 June 2017 # Springer-Verlag GmbH Germany 2017
Abstract Background Peripherally inserted central catheters (PICCs) represent a mainstay of intravascular access in the neonatal intensive care setting when long-term vascular access is needed. Ideally, PICCs should be inserted and maintained in a central position with the tip ending in the superior or inferior vena cava. This is not always achievable, and sometimes the tip remains in a peripheral location. Higher complication rates have been reported with non-central PICCs; however these findings have not been confirmed in a solely neonatal series and PICCs with tips in peripheral veins have not been studied. Objective To compare complication rates and length of catheter duration related to PICC position in neonates.
* Bernard Goldwasser
[email protected]
1
Department of Radiology, Jacobi Medical Center/Albert Einstein College of Medicine, 1400 Pelham Parkway South, Building 1, Room 4N15, Bronx, NY 10461, USA
2
Department of Neonatology, Sheridan Healthcare of Texas, Southlake, TX, USA
3
Department of Epidemiology & Population, Albert Einstein College of Medicine, Jack and Pearl Resnick Campus, Bronx, NY, USA
4
Departments of Pediatrics and Radiology, Albert Einstein College of Medicine, Bronx, NY, USA
5
Department of Radiology, Montefiore Medical Center, Bronx, NY, USA
6
Division of Neonatology, Department of Pediatrics, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, NY, USA
Materials and methods We conducted a retrospective analysis of all PICCs inserted in term and preterm infants in a tertiary neonatal intensive care unit between May 2007 and December 2009. A single pediatric radiologist reinterpreted the catheter tip site on initial anteroposterior (AP) chest radiographs and categorized sites as central (superior vena cava, inferior vena cava, brachiocephalic vein), intermediate (subclavian, axillary, common or external iliac veins), or peripheral (veins peripheral to axillary or external iliac veins). We analyzed complication rates and length of catheter duration among the three categories. Results We collected data on a total of 176 PICCs. Infants with PICCs in a central location had a significantly lower complication rate (18/97, 19%) than those with the PICC tip in an intermediate (24/64, 38%) or peripheral (9/15, 60%) locations (P=0.0003). Length of catheter duration was noted to be longest with central, intermediate with intermediate, and shortest with peripheral PICC tip locations (17.7±14.8 days for central vs. 11.4±10.7 days for intermediate vs. 5.4±2.5 days for peripheral, P=0.0003). Conclusion A central location is ideal for the tip of a PICC. When this is not achievable, an intermediate location is preferable to a more peripheral position. Keywords Chest radiography . Intravascular access . Neonatal intensive care unit . Neonates . Peripherally inserted central catheter
Introduction Peripherally inserted central catheters (PICCs) represent a mainstay of intravascular access in the neonatal intensive care setting when long-term vascular access is needed. Ideally, PICCs should be inserted and maintained in a
Pediatr Radiol
central position with the tip ending in the superior or inferior vena cava [1–3]. However this is not always achievable because the catheter sometimes cannot be advanced, is inserted too deep and then pulled back too far, or migrates. When a catheter cannot be advanced, it is presumed to be caused by venospasm or the anatomy of the vessel, including tortuosity of the vessel or the presence of valves [1, 2]. While a PICC is meant by definition to be used as a central venous catheter, when the above situations occur, this type of catheter has in practice been used until a new PICC with a more ideal tip location is placed or until it is no longer useable or needed. Evidence from studies involving older children suggests that PICCs left in a non-central position might lead to increased risk of complications such as infiltration, infection and phlebitis [1–5]. Therefore, in practice, the use of PICCs whose tips are non-central is only acceptable when intravascular access is absolutely essential and no better alternatives are available. Studies addressing safety related to PICC tip location include adult and pediatric populations [1–3, 6], although studies exclusively evaluating neonatal PICCs are rare [3, 7]. The purpose of this study was to compare complication rates and length of catheter duration related to PICC tip location exclusively in neonates. We evaluated the impact on complications for variables including gestational age, birth weight and site of insertion. Additionally, we studied the relationship between gestational age and success rate of central placement.
Materials and methods Our institutional review board approved this retrospective chart review, which was conducted in a level III neonatal intensive care unit (NICU) in Bronx, NY. We included newborns who required PICC placement from May 1, 2007, to Dec. 31, 2009. We excluded infants with PICCs not visualized on radiograph or lacking available radiographs for placement confirmation. Single-lumen PICCs (Vygon silicone 2 F or Argyle 1.9 F) were inserted using an introducer that is removable following PICC placement by a physician, nurse practitioner, or physician assistant by direct vein visualization or palpation. Neither ultrasound nor fluoroscopic guidance was used. none of the catheters were trimmed. When the native lines were too long, the extra catheter length was left coiled outside the patient under an occlusive dressing. Heparin (0.5 U/mL) was continuously infused at 1 mL/h immediately after all catheter placements and only solution concentrations of 12.5% dextrose or less and an osmolality of <1,000 milliosmoles (mOsm)/L were used for non-central PICCs. The final position was determined on portable anteroposterior (AP) chest
radiographs. When an AP radiograph could not precisely identify the tip position, lateral views were obtained or contrast agent (Omnipaque 200, <0.3 mL, depending on priming volume as indicated in specific PICC product information) was injected to opacify the PICC to aid in visualization. A single fellowship-trained pediatric radiologist (B.H.T., with 12 years of experience) who was blinded to the outcome reinterpreted the radiologic studies retrospectively. We used the first radiograph with the PICC in place to determine position. We categorized the PICC position as central (central-PICC) when the tip was located in the superior vena cava (SVC), inferior vena cava (IVC), or brachiocephalic vein (left or right); intermediate (intermediate-PICC) when the tip was in the subclavian, axillary, common iliac or external iliac vein; or peripheral (peripheral-PICC) when the tip was in a more peripheral vein (Fig. 1). The PICC tip location was categorized as follows: axillary, or between the edge of the first rib and the medial portion of the humeral head; subclavian, or medial to the first rib until the confluence with the brachiocephalic vein (when the catheter turns inferiorly); brachiocephalic (for both left and right), when the catheter turns inferiorly until within 1 cm of the cardiac silhouette for preterm and 2 cm for term infants; SVC, or above the described brachiocephalic border until the cardiac silhouette; common/external iliac, or between L4–5 intervertebral disc space and the superior edge of the superior pubic ramus; IVC, or L4–5 intervertebral disc space until the level of the diaphragm. We analyzed infant characteristics (birth weight, gestational age, days of age at the time of catheter placement), site of insertion, initial location of tip, length of catheter duration, reason for removal and associated complications. We then categorized each PICC removal as elective or due to complication. Complications included extravasation of fluid and swelling (with erythema and serous drainage without changes in white blood cell count, Creactive protein or other signs of infection that prompted use of antibiotics), phlebitis (redness or swelling at the site of insertion, with laboratory changes concerning for infection and therefore treated with antibiotics) with or without thrombus, central line associated bloodstream infection (defined as one or more positive blood cultures within 48 h of catheter insertion) or mechanical (leaking, occlusion, cracked catheter hub). There were no complications related to renal vein thrombosis, which has a reported association with central venous catheters [8, 9]. All of the complications required catheter removal. We performed statistical analysis using the χ2 test or Fisher exact test for categorical variables and the Student’s t-test and analysis of variance (ANOVA) for continuous variables. P-values <0.05 were considered
Pediatr Radiol
Fig. 1 Annotated anteroposterior radiographs of the chest (a) and abdomen/pelvis (b). Annotated chest radiograph (a) demonstrates the expected course of a correctly placed right upper extremity PICC (white traced line terminating at the superior cavoatrial junction). The tip could potentially terminate at any of the following categories of locations: central-PICC: central to the black line or central to the junction of the subclavian and brachiocephalic vein; intermediate-PICC: between the light gray and black lines or between the junction of the axillary and peripheral veins and the junction of the subclavian and brachiocephalic vein; peripheral-PICC: peripheral to the light gray line or peripheral to the junction of the axillary and peripheral veins (dark gray line, junction of
the subclavian and axillary veins). Annotated abdomen/pelvis radiograph (b) demonstrates the expected course of a correctly placed right lower extremity PICC (white traced line terminating at the IVC). The tip could potentially terminate at any of the following categories of locations: central-PICC: superior to the black line or superior to the junction of the common iliac vein and IVC; intermediate-PICC: between the gray and black lines or between the junction of the external iliac and femoral veins and the junction of the common iliac vein and IVC; peripheralPICC: inferior to the gray line or inferior to the junction of the external iliac and femoral veins. Incidentally noted is a right pneumothorax. IVC inferior vena cava, PICC peripherally inserted central catheter
statistically significant. Non-normally distributed continuous variables were rank transformed prior to analysis. Logistic regression models were fit to control for birth weight and gestational age.
on or unavailability of radiographs. Of the included PICCs, 97 (55%) PICC tips were placed in central position, 64 (36%) in intermediate position and 15 (9%) in peripheral position. Infants with central-PICCs had a significantly lower complication rate (18/97, 19%) than those with intermediate-PICCs (24/64, 38%) or peripheralPICCs (9/15, 60%). The P-value was <0.001 (P<0.05 is significant, but for a larger P-value it is unclear whether it is really nonsignificant or related to the small sample size). After controlling for gestational age and birth weight, the odds of having complications were 2.8-fold higher in intermediate-PICCs and 8.8-fold higher in peripheral-PICCs compared to central-PICCs (Table 2).
Results A total of 155 newborns, mean gestational age of 29.7 ±4.4 weeks (23–41 weeks), and mean birth weight of 1,361±818 g (461–3,835 g), required 176 PICCs during the study period and met the inclusion criteria (Table 1). Four PICCs were excluded because of non-visualization
Table 1 Patient demographics Gestational age at birth (weeks) 33–41 23–32 Birth weight (grams) 0–1,500 1,501+
PICCs with complications (#)
Total PICCs
PICCs with complications (%)
13 38
39 137
33 28
40 11
130 46
31 24
PICC peripherally inserted central catheter
Pediatr Radiol Table 2 Complication rate of intermediate vs. central and peripheral vs. central PICC tip location
Without controlling for GA and BW
Controlling for GA and BW
PICC tip location
Odds ratio
P-value*
95% CI
Odds ratio
P-value*
Intermediate vs. central
2.63
0.008
1.28–5.41
2.83
0.006
95% CI 1.35–5.91
Peripheral vs. central
6.58
0.001
2.08–20.85
8.80
0.0007
2.49–31.10
BW birth weight, CI confidence interval, GA gestational age, PICC peripherally inserted central catheter *P-value <0.05 considered statistically significant
Gestational age did not have an effect on the complication rate. Using a definition of less than 32 weeks as very premature, there was no significant difference between complication rates in very premature compared to preterm/term infants (28% vs. 33%, P=0.69). Additionally, birth weight did not have an effect on complication rate. Using a cutoff of 1,500 g, 31% of infants with lower birth weight were found to have complications, vs. 24% of those with higher birth weight (P=0.45). Very premature infants were found to have a higher likelihood of successful placement of the PICC tip in a central location compared to the premature/term infant group (59% vs. 41%, P=0.0083) and were much less likely to have a peripheral placement (5% vs. 20%, P=0.0083, Table 3). A slightly higher complication rate was found in PICCs placed in the upper extremity when compared to those placed in the lower extremity, although this did not meet statistical significance (P=0.27, Table 4). There was no difference in complication rates between PICCs with tips in the brachiocephalic vein (3/16, 19%) and those with tips in the SVC (12/55, 22%, P=1.0) or IVC (3/26, 12%, P=0.66). Central-PICCs were noted to have a longer mean length of catheter duration compared to intermediate-PICCs, which had a longer mean length of catheter duration compared to peripheral-PICCs (17.7±14.8 days for central-PICCs vs. 11.4 ±10.7 days for intermediate-PICCs vs. 5.4±2.5 days for peripheral-PICCs, P<0.001). When analyzing only PICCs with complications to determine length of catheter duration, similar means were demonstrated (16.3, 11.3 and 5.1 days for centralPICCs, intermediate-PICCs and peripheral-PICCs respectively). Kaplan–Meier analyses of PICCs with complications and all PICCs showed a shorter length of catheter duration with non-central compared to central PICCs (Fig. 2).
Table 3
Discussion PICC placement with the tip in the superior or inferior vena cava has been accepted as an ideal central position in a broad age range of pediatric patients [1, 2]. Another study demonstrated this as well with exclusively premature lowbirth-weight neonates (gestational age at insertion 27– 31 weeks/birth weight 760–1,170 g) [3]. The PICC tip location categories in that study were different from ours. They limited the definition of central location to the SVC or IVC, while we included brachiocephalic because we found no difference in complication rate between PICC tips in the brachiocephalic vein when compared to the SVC or IVC, a finding that was also demonstrated in their study [3]. They categorized brachiocephalic, subclavian, common iliac and external iliac veins as non-central, while we included the latter three sites as an intermediate category of preference (intermediate-PICCs). Our least preferred sites, categorized as peripheral-PICCs in our study, were called short catheters and were not analyzed in their study because of their similarity to peripheral intravenous lines. Our study addresses peripheral-PICCs, as defined in the methods, because in our experience it is necessary at times to use a peripheral-PICC for life-saving medications and fluid until more optimal access can be obtained. Our findings added that intermediate-PICCs, while less preferable than central-PICCs, are significantly more preferable to PICCs with tips peripheral to the axillary or external iliac veins. The complication rate for peripheralPICCs was ~200% greater compared to central-PICCs, whereas intermediate-PICCs had only ~100% increase in complications compared to central-PICCs. PeripheralPICCs had a 60% increase in complications when compared to intermediate-PICCs. Additionally, whereas the length of catheter duration for intermediate-PICCs was only 64% that of central-PICCs, the length of catheter
PICC tip location by gestational age (GA)
Tip location
≤32 weeks
>32 weeks
Central Intermediate Peripheral Total
81 (59%) 49 (36%) 7 (5%) 137
16 (41%) 15 (38%) 8 (20%) 39
Table 4
Complication rate by site of insertion
Insertion site
Complication
Elective removal
Upper extremity Lower extremity Scalp
38/129 (29%) 11/44 (25%) 1/3 (33%)
91/129 (71%) 33/44 (75%) 2/3 (67%)
Pediatr Radiol
b
PICCs Removed Due to Complication or End of Study
PICCs Removed Due to Complication
1
1
0.9
0.9
0.8
0.8 PICC Survival Probability
PICC Survival Probability
a
0.7 0.6 0.5 0.4
0.3
0.7 0.6 0.5 0.4
0.3
0.2
0.2
0.1
0.1
0
0
10
20
30
Central
40
50 Days Intermediate
60
70
80
90
100
Peripheral
0
0
10
20
30
Central
Days Intermediate
40
50
60
70
Peripheral
Fig. 2 Kaplan–Meier survival analysis demonstrates that catheters whose tips are in a peripheral location have the lowest survival probability, and catheters whose tips are in a central location have the
highest survival probability when studying all PICCs removed (a) and when studying those only removed because of complications (b). PICC peripherally inserted central catheter
duration for peripheral-PICCs was only 31% that of central-PICCs. We performed a separate analysis including only PICCs removed secondary to complications and compared to the analysis of all PICCs in order to avoid confounding data, where a peripheral-PICC was removed not following a complication but following placement of a more ideally located central-PICC. The mean length of catheter duration for this subset was similar to the mean including all PICCs. Very premature infants were more likely to have successful placement of a central-PICC compared to infants in the premature/term group. A possible explanation for this finding is that larger neonates are likely to be complex surgical cases who have had multiple peripheral intravenous (IV) access attempts, thereby limiting the number of available veins. Very premature infants typically have umbilical catheters placed for the first week and have their PICC placed without ever having a peripheral IV attempt, thus resulting in many more options for PICC insertion sites. Additionally, very premature infants have more visible veins because of having less subcutaneous fat, and as ultrasound was not utilized, they had more available veins for PICC placement attempts. Finally, similar to previous, much larger studies [7, 10], we found that site of insertion (upper vs. lower extremity) did not significantly change the complication rate. Of note, a different study [11] demonstrated a higher complication rate with upper extremity PICCs when used for total parenteral nutrition. Specifically, there was a significantly increased rate of catheterassociated bloodstream infections with coagulasenegative Staphylococcus, as well as a significantly increased rate of cholestasis. The time to complication
and the length of catheter duration were also significantly more favorable in the lower extremity PICCs. The latter study, while larger than our study, was not as large as the two more recent ones [7, 10]. A limitation in our study was that for upper extremity PICCs, the specific vein in which the PICC was inserted was not recorded and therefore could not be controlled for. It is possible that the cephalic vein was a preferred site of insertion because this site is farther from the brachial artery. This might have had an impact on the success rate of central tip placement because the route to the axillary vein via the cephalic vein takes a sharp turn compared to the direct continuation of the basilic or brachial vein into the axillary vein. This bias might account for the higher success rate of central tip placement in lower extremity PICCs (although statistical significance was not demonstrated). Another limitation was the small sample size in the peripheral-PICC group. There were few peripheral-PICCs because they anecdotally had a higher complication rate and were therefore considered suboptimal. Future study looking exclusively at peripheral-PICCs over a longer study period might demonstrate the findings of our study with greater statistical power. Additionally, as stated elsewhere [3], peripheral-PICCs might be seen as more similar to peripheral IVs. We did not compare the complication rate and length of catheter duration of the peripheral-PICCs to peripheral IVs, which might be an interest for further study. Finally, the type of infusate for each individual PICC was not controlled for, and because different infusates cause different degrees of irritation and endothelial injury, PICCs in which more irritating infusates were used might have had higher complication rates, irrespective of the PICC tip location.
Pediatr Radiol
Conclusion
3.
Central location of PICCs includes the SVC, IVC and brachiocephalic veins and is the ideal location for the tip of the PICC. Central-PICCs have fewer complication rates and longer length of catheter duration. When this is not achievable, an intermediate location, defined as subclavian and axillary veins in the upper extremity or common iliac and external iliac veins in the lower extremity, is preferable to more peripherally located PICC tips.
4.
5. 6.
7. Compliance with ethical standards Conflicts of interest None
8.
9.
References 10. 1.
2.
Racadio J, Doellman D, Johnson N et al (2001) Pediatric peripherally inserted central catheters: complication rates related to catheter tip location. Pediatrics 107:e28 Jumani K, Advani S, Reich NG et al (2013) Risk factors for peripherally inserted central venous catheter complications in children. JAMA Pediatr 167:429–435
11.
Jain A, Deshpande P, Shah P (2013) Peripherally inserted central catheter tip position and risk of associated complications in neonates. J Perinatol 33:307–312 Orme RMLE, McSwiney MM, Chamberlain-Weber RFO (2007) Fatal cardiac tamponade as a result of a peripherally inserted central venous catheter: a case report and review of the literature. Brit J Anaesth 99:384–388 Hogan MJ (1999) Neonatal vascular catheters and their complications. Radiol Clin N Am 37:1109–1125 Sharp R, Esterman A, McCutcheon H et al (2014) The safety and efficacy of midlines compared to peripherally inserted central catheters for adult cystic fibrosis patients: a retrospective, observational study. Int J Nurs Stud 51:694–702 Bashir RA, Kamala S, Vayalthrikkovil S et al (2016) Association between peripherally inserted central venous catheter insertion site and complication rates in preterm infants. Am J Perinatol 33:945–950 Yang C-Y, Fu R-H, Lien R, Yang P-H (2014) Unusual presentation of renal vein thrombosis in a preterm infant. Urol Case Rep 2:117–119 Kuhle S, Massicotte P, Chan A, Mitchell L (2004) A case series of 72 neonates with renal vein thrombosis. Data from the 1-800-NOCLOTS Registry. Thromb Haemost 92:729–733 Wrightson DD (2013) Peripherally inserted central catheter complications in neonates with upper versus lower extremity insertion sites. Adv Neonatal Care 13:198–204 Hoang V, Sills J, Chandler M et al (2008) Percutaneously inserted central catheter for total parenteral nutrition in neonates: complications rates related to upper versus lower extremity insertion. Pediatrics 121:e1152–e1159